The Impact of Environmental Variability on Fishers’ Harvest Decisions in Chile

.title[
# The Impact of Environmental Variability on Fishers’ Harvest Decisions in Chile
]
.subtitle[
## Using a Multi-Species Approach
]
.author[
### Felipe J. Quezada-Escalona
]
.date[
### Encuentro EfD Chile 2025
]

---

<div>
<style type="text/css">.xaringan-extra-logo {
width: 110px;
height: 128px;
z-index: 0;
background-image: url(https://fquezadae.github.io/Slides-Econometria/figs/depto_economia_blanco.png);
background-size: contain;
background-repeat: no-repeat;
position: absolute;
top:1em;right:1em;
}
</style>
<script>(function () {
  let tries = 0
  function addLogo () {
    if (typeof slideshow === 'undefined') {
      tries += 1
      if (tries < 10) {
        setTimeout(addLogo, 100)
      }
    } else {
      document.querySelectorAll('.remark-slide-content:not(.title-slide):not(.inverse):not(.hide_logo)')
        .forEach(function (slide) {
          const logo = document.createElement('div')
          logo.classList = 'xaringan-extra-logo'
          logo.href = null
          slide.appendChild(logo)
        })
    }
  }
  document.addEventListener('DOMContentLoaded', addLogo)
})()</script>
</div>

# Introduction

.pull-left[
- Marine resource distribution and abundance is changing due to climate variability, with heterogenous spatial effects <a name=cite-Poloczanska2013-qq></a><a name=cite-sumaila2011></a>([Poloczanska, Brown, Sydeman et al., 2013](#bib-Poloczanska2013-qq); [Sumaila, Cheung, Lam et al., 2011](#bib-sumaila2011)).

- Harvest levels would be affected <a name=cite-Quezada2023-hc></a>([Quezada, Tommasi, Frawley et al., 2023](#bib-Quezada2023-hc)), 
  - Prices and value of catches, fishing costs, fishers’ incomes, among others ([Sumaila, Cheung, Lam et al., 2011](#bib-sumaila2011))
]

.pull-right[
<img src="figs/catch_potential.jpeg" style="width:100%;max-width:100%;height:auto;display:block;margin:0 auto;">

<span style="font-size:0.55em; display:block; text-align:right; margin-top:4px;">
**Source:** <a name=cite-Cheung2010></a>[Cheung, Lam, Sarmiento et al. (2010)](#bib-Cheung2010)
</span>

]

???

Recordar que uno apreta C para clonar, P para Presenter View, H para tener un mapa de las teclas

---

# Introduction

## Research question

How will fishing decisions, aggregate catch levels, and the price of marine resources be affected under different climatic scenarios in the multispecies small pelagic fishery (SPF) in Chile?

- How do fishers **substitute between species**?
  - Contribute to the limited local literature on multi-species economic modeling and climate variability in Chile (e.g., <a name=cite-Pena-Torres2017-gn></a>[Peña-Torres, Dresdner, and Vasquez (2017)](#bib-Pena-Torres2017-gn))
  - Understand fishers’ adaptive capacity helps to inform climate-resilient fisheries policies
  - We will focus in **climate variability** to estimate short-run responses.
    - *i.e.*, climate change effect without adaptation <a name=cite-auffhammer2018></a>([Auffhammer, 2018](#bib-auffhammer2018))
    - Upper-bound estimates

---

# Chile’s Small Pelagic Fishery

---

# Chile’s Small Pelagic Fishery (SPF)

## Some facts

- Mainly composed by anchoveta, Jack mackerel, Sardine
- ~94% of national catch <a name=cite-SUBPESCA2020></a>([SUBPESCA, 2020](https://www.subpesca.cl/portal/618/articles-106845_documento.pdf))
- Mostly harvested with purse-seiners
- In the Central-South (CS) region (Valparaiso-Los Lagos) all three species play a major role.
- SPF have been used primarily for fishmeal and fish oil production ([Peña-Torres, Dresdner, and Vasquez, 2017](#bib-Pena-Torres2017-gn)) (~85% of jack mackerel for reduction)
  
## Regulations
- Quota (TAC); divided between the small-scale and indutrial sector
- Industrial sector operates under ITQ
- RAE (*Regimen Artesanal de Extracción*) in some areas for small-scale fishery
  - Allocates regional quota to area or fishermen organization (i.e., catch shares)
- Anchoveta and sardine are regulated as a mixed-species fishery
  
---

# Chile’s Small Pelagic Fishery (SPF)

## Status of the stocks (CS)

- Anchoveta: 
  - Collapsed until 2018, 
  - Overexploited in 2019, 
  - Since 2020, within MSY limits.

- Sardine:
  - Within MSY levels, except in 2021 and 2023 (overexploited)

- Jack mackerel:
  - Overexploited until 2018, then within MSY limits. 
    
---

# Methods and data

---

# Methodology Overview

Based on <a name=cite-Kasperski2015-jm></a>[Kasperski (2015)](#bib-Kasperski2015-jm):

- Econometrics models
  - Estimate stock dynamics
  - Estimate trip-level costs
  - Estimate annual trips
  - Estimate inverse demand

- Use parameters to:
  - Obtain the optimal **harvest** and **quota** conditional on climate variables.
  - Simulate the effects on harvest, prices and **profits**, and **species substitution** of different climate scenarios.

---

# Data Sources

## Harvest and biomass data

From IFOP (2012-2025):
- Trip-level data:
  - ID, departure and arrival times, vessel capacity, fleet and gear type, ports of departure and landing, haul timing and location, species, retained catch. 
- Annual stock biomass by macro region and species (e.g. south-central Chile) -- (1999-2025)
- Monthly landings by port/species.  
- Prices paid by processing plants (IFOP surveys; month-region)

---

# Data Sources

## Environmental covariates

- Daily salinity, sea surface temperature, and current speed and direction <a name=cite-GLORYS12V1></a>([E.U. Copernicus Marine Service Information, 2025a](https://doi.org/10.48670/moi-00021))
- Hourly wind speed and direction at the surface <a name=cite-WIND_GLO_PHY></a>([E.U. Copernicus Marine Service Information, 2025b](https://doi.org/10.48670/moi-00185))
- Chlorophyll-a concentration <a name=cite-GlobColour></a>([E.U. Copernicus Marine Service Information, 2025c](https://doi.org/10.48670/moi-00281))
]

.pull-right[
<img src="figs/env_data_map.svg" style="width:100%; max-width:100%; height:auto; display:block; margin:0 auto;">
]

---

# Data Sources

## To be requested

- Average wage pay to crew member per hour (available?)
- Diesel cost.
- Permits by vessels
- Quota prices (auction or secondary markets, if available)
  - Captures forward-looking behavior and information <a name=cite-Birkenbach2024></a>([Birkenbach, Lee, and Smith, 2024](https://doi.org/10.1086/727356)).
  - Simplify the dynamic model to a static one <a name=cite-reimer2022structural></a>([Reimer, Abbott, and Haynie, 2022](#bib-reimer2022structural)).
- Quota by area/fishing organization for small-scale sector, and ITQ for industrial (by vessel?)

---

# Data Sources

## Data for projections

**Bio-ORACLE**

- Advantages:
  - Different climate scenarios
  - SST, salinity, currents and chlorophyll (4km resolution)

- Disvantages
  - Only decadal (e.g., 2040–2050) projections
  - No winds; CMIP6 for winds? (~100 km).
]

.pull-right[
  <img src="figs/oracle.png" style="width:100%; max-width:100%; height:auto; display:block; margin:0 auto;">
]

---

# Econometrics models

---

# Model 1: Stock Dynamics

`$$x_{i,y+1} + h_{iy} = \underbrace{(1 + r_i)x_{iy} + \eta_i x_{iy}^2}_{R_i(x_{iy})} + \underbrace{\sum_{j \neq i}^{n-1} a_{ij} x_{iy} x_{jy}}_{I_i(x_y)} + \rho_i Env_{iy} + \varepsilon_{iy} \quad i=1,\ldots,n$$`
where: 
- `$x_{iy}$` is the fish stock by species `$i=1,\ldots,n$` in year `$y$`, `$n$` is the total number of species, 
- `$h_{iy}$` is the annual harvest of species `$i$` on year `$y$`, 
- `$r_i$` is the intrinsic growth rate of the resource `$i$`,
- `$\eta_i$` is a density-dependent factor related to the carrying capacity, 
- `$\alpha_{ij}$` are the interaction parameters between species. 
- `$Env_{iy}$` includes year–macroregion averages of **environmental covariates** (SST and chlorophyll-a).

The system of `$n$` growth equations can be estimated simultaneously using SUR

---

# Model 2: Trip-Level Costs

`$$C_{vg} = \sum_{i=1}^{2n+M+k} \alpha_{g, \mathbf{X}_i} \mathbf{X}_{ivg} + \frac{1}{2} \sum_{i=1}^{2n+M+k} \sum_{j=1}^{2n+M+k} \alpha_{g, \mathbf{X}_i\mathbf{X}_j} \mathbf{X}_{ivg} \mathbf{X}_{jvg}$$`

where `$C_{vg}= w z_{vg}^*$` is the total cost incurred by vessel `$v=1,\ldots,V_g$` conditional on gear used `$g=1,\ldots,G$`  -- Mostly purse seiners!
- `$w$` is a matrix of variable input prices.
- `$z_{vg}^*$` is the optimal quantity of input used
  - Crew members?
  - Time spent at sea?
  - Distance traveled?
- `$\mathbf{X}^{'}_{vg} = [w;h_{vg};x;Z_v;Env]$` is a matrix of explanatory variables:
  - `$Env$` is a matrix of **environmental covariates** (e.g., wind intensity and wave conditions in each trip)

---

# Model 3: Total Annual Trips

The number of trips a vessel will take in a given year is assumed to follow a Poison distribution: 
`$$Pr\left[T^{*}_{vgy} = t_v\right] = \frac{exp^{-exp(U^{'}_{vg}\beta_g)}exp(U^{'}_{vg}\beta_g)^{t_v}}{t_v !}$$`
where `$U_{vg}=[p,w,h_{vg},\bar{q},Z_{vg}, Env]$` is a matrix of explanatory variables,
- `$t_v$` is the number of trips taken by vessel `$v$` using gear type `$g$` in year `$y$`
- `$Env$` include variables that reflect **annual weather conditions**
  - Accumulation of *bad weather days*? 
  - *Number of storms*?

---

# Model 4: Inverse Demand

The price of each species is modeled using an Inverse Almost Ideal Demand System (IAIDS). The log of the price `$p_{iy}$` of a species `$i$` in year `$y$` is the following:

`$$\ln p_{iy} = \sum_j^n \gamma_j \ln h_{j,y} + \gamma_{H} \ln H_y + \gamma_{FM} \ln P^\text{FishMeal}_y + \epsilon_{iy}$$`
where:
- `$H_y = \sum h_{j,y}$`
- `$P^\text{FishMeal}_y$` is the fish meal world price

Harvest may be endogenous 
  - Three Stage Least Squares (3SLS) procedure, 
  - `$h_{j,y}$` instrumented by variables that affect supply function such as *SST*, *Chl*, and fuel prices.
  
---

# Preliminary results

---

# Is there any subtitution?

## Small-scale vessels

---

# Is there any subtitution?

## Industrial vessels

---

# Stock dynamics

- SST and CHL improves model performance (F = 1.908; `$p$`-value = 0.07).

???

- Herrick, et al. (2009): "Sardine is known to be more productive during warm-water regimes in the California Current ecosystem."

---

# What’s Next?

- Finish biomass estimations
- Start soon with total annual trips
  - Hopefully, a short paper can come out of that work
- Two undergraduate students are working on the inverse demand (i.e., price) module for their theses
  - Results expected by July 2026
  - If time allows, they will also analyze long-run and short-run dynamics using a VEC model
  - Plan to write a paper based on their dissertations

---

# ¡Muchas gracias!

<span style="color:#f59f18; font-size:1.3em; font-weight:bold;">¿Preguntas?</span>

**Felipe J. Quezada-Escalona**

<a href="https://felipequezada.com" target="_blank" style="
  font-size:1em;
  background:linear-gradient(#f59f18);
  -webkit-background-clip:text;
  -webkit-text-fill-color:transparent;
  text-decoration:none;
">
🌐 felipequezada.com
</a>

<br>

<span style="font-size:0.8em; color:lightgray;">
I gratefully acknowledge financial support for this research from ANID-Chile, under project Fondecyt Iniciación No. 11250223.
</span>

---

# References

<a name=bib-auffhammer2018></a>[Auffhammer, M.](#cite-auffhammer2018)
(2018). "Quantifying economic damages from climate change". In:
_Journal of Economic Perspectives_ 32.4, pp. 33-52.

<a name=bib-Birkenbach2024></a>[Birkenbach, A. M., M. Lee, and M. D.
Smith](#cite-Birkenbach2024) (2024). "Counterfactual Modeling of
Multispecies Fisheries Outcomes under Market-Based Regulation". In:
_Journal of the Association of Environmental and Resource Economists_
11.3, pp. 755-796. DOI:
[10.1086/727356](https://doi.org/10.1086%2F727356). eprint:
https://doi.org/10.1086/727356.

<a name=bib-Cheung2010></a>[Cheung, W. W., V. W. Lam, J. L. Sarmiento,
et al.](#cite-Cheung2010) (2010). "Large-scale redistribution of
maximum fisheries catch potential in the global ocean under climate
change". In: _Global change biology_ 16.1, pp. 24-35.

<a name=bib-GlobColour></a>[E.U. Copernicus Marine Service
Information](#cite-GlobColour) (2025c). _Global Ocean Colour
(Copernicus-GlobColour)_. Accessed on 23-SEP-2025. DOI:
[10.48670/moi-00281](https://doi.org/10.48670%2Fmoi-00281).

<a name=bib-WIND_GLO_PHY></a>[E.U. Copernicus Marine Service
Information](#cite-WIND_GLO_PHY) (2025b). _Global Ocean Hourly
Reprocessed Sea Surface Wind and Stress from Scatterometer and Model_.
Accessed on 23-SEP-2025. DOI:
[10.48670/moi-00185](https://doi.org/10.48670%2Fmoi-00185).

---

# References

<a name=bib-GLORYS12V1></a>[E.U. Copernicus Marine Service
Information](#cite-GLORYS12V1) (2025a). _Global Ocean Physics
Reanalysis_. Accessed on 23-SEP-2025. DOI:
[10.48670/moi-00021](https://doi.org/10.48670%2Fmoi-00021).

<a name=bib-Finkbeiner2015-bs></a>[Finkbeiner, E.
M.](#cite-Finkbeiner2015-bs) (2015). "The role of diversification in
dynamic small-scale fisheries: Lessons from Baja California Sur,
Mexico". In: _Glob. Environ. Change_ 32, pp. 139-152.

<a name=bib-Gonzalez-Mon2021-kj></a>[Gonzalez-Mon, B., Ö. Bodin, E.
Lindkvist, et al.](#cite-Gonzalez-Mon2021-kj) (2021). "Spatial
diversification as a mechanism to adapt to environmental changes in
small-scale fisheries". In: _Environ. Sci. Policy_ 116, pp. 246-257.

<a name=bib-Jardine2020-um></a>[Jardine, S. L., M. C. Fisher, S. K.
Moore, et al.](#cite-Jardine2020-um) (2020). "Inequality in the
Economic Impacts from Climate Shocks in Fisheries: The Case of Harmful
Algal Blooms". In: _Ecol. Econ._ 176, p. 106691.

<a name=bib-Kasperski2015-jm></a>[Kasperski,
S.](#cite-Kasperski2015-jm) (2015). "Optimal Multi-species Harvesting
in Ecologically and Economically Interdependent Fisheries". In:
_Environ. Resour. Econ._ 61.4, pp. 517-557.

---

# References

<a name=bib-Kasperski2013-jz></a>[Kasperski, S. and D. S.
Holland](#cite-Kasperski2013-jz) (2013). "Income diversification and
risk for fishermen". En. In: _Proc. Natl. Acad. Sci. U. S. A._ 110.6,
pp. 2076-2081.

<a name=bib-Pena-Torres2017-gn></a>[Peña-Torres, J., J. Dresdner, and
F. Vasquez](#cite-Pena-Torres2017-gn) (2017). "El Niño and Fishing
Location Decisions: The Chilean Straddling Jack Mackerel Fishery". In:
_Mar. Resour. Econ._ 32.3, pp. 249-275.

<a name=bib-Poloczanska2013-qq></a>[Poloczanska, E. S., C. J. Brown, W.
J. Sydeman, et al.](#cite-Poloczanska2013-qq) (2013). "Global imprint
of climate change on marine life". En. In: _Nat. Clim. Chang._ 3.10,
pp. 919-925.

<a name=bib-Powell2022-wj></a>[Powell, F., A. Levine, and L.
Ordonez-Gauger](#cite-Powell2022-wj) (2022). "Climate adaptation in the
market squid fishery: fishermen responses to past variability
associated with El Niño Southern Oscillation cycles inform our
understanding of adaptive capacity in the face of future climate
change". En. In: _Clim. Change_ 173.1-2, p. 1.

<a name=bib-Quezada2023-hc></a>[Quezada, F. J., D. Tommasi, T. H.
Frawley, et al.](#cite-Quezada2023-hc) (2023). "Catch as catch can:
markets, availability, and fishery closures drive distinct responses
among the U.S. West Coast coastal pelagic species fleet segments". En.
In: _Can. J. Fish. Aquat. Sci._. Just-IN.

---

# References

<a name=bib-reimer2022structural></a>[Reimer, M. N., J. K. Abbott, and
A. C. Haynie](#cite-reimer2022structural) (2022). "Structural
behavioral models for rights-based fisheries". In: _Resource and Energy
Economics_ 68, p. 101294.

<a name=bib-SUBPESCA2020></a>[SUBPESCA](#cite-SUBPESCA2020) (2020).
_Informe Sectorial de Pesca y Acuicultura 2019_. Accessed: 02-04-2025.
URL:
[https://www.subpesca.cl/portal/618/articles-106845_documento.pdf](https://www.subpesca.cl/portal/618/articles-106845_documento.pdf).

<a name=bib-sumaila2011></a>[Sumaila, U. R., W. W. Cheung, V. W. Lam,
et al.](#cite-sumaila2011) (2011). "Climate change impacts on the
biophysics and economics of world fisheries". In: _Nature climate
change_ 1.9, pp. 449-456.

<a name=bib-Zhang2011-wv></a>[Zhang, J. and M. D.
Smith](#cite-Zhang2011-wv) (2011). "Heterogeneous Response to Marine
Reserve Formation: A Sorting Model approach". In: _Environ. Resour.
Econ._ 49.3, pp. 311-325.

---

# Appendix A: Numerical optimization

---

# Appendix A: Numerical optimization

## Vessel maximization problem

In each year, a vessel maximizes profits by choosing their optimal number of
trips `$T_g$` and harvest levels per trip `$h_{g\tau}$` given a gear type:
`$$\begin{align*}
\max_{h_{gt}, T_g} \quad \pi_{vgt} & = \sum_{\tau=t}^{T_g} \rho^\tau \left\{ P(h) h_{g\tau} - C_g(h_{g\tau} | w, x, Z, Env) \right\} \quad \tau = t,\ldots, T_g \nonumber \\
\textbf{s.t} \quad q_{g,t+1} & = \omega \ast \bar{q} - \sum_{t=1}^{t} h_{gt} \geq 0, \quad t = 1, \dots, T-1, \quad g = 1, \dots, G
\end{align*}$$`

- where:
  - `$\rho$` is the intra-annual discount factor, 
  - `$\omega$` is a vector of shares of `$\bar{q}$`, and 
  - `$h_{lt}=0$` for all `$l\neq g$`.

???

- The vector of shares is obtained from historical data on harvest. 
- The optimal profit from the maximization problem is `$\pi_{vgy}^* (p,w,x,Z,\bar{q},\omega, Env)$`, 
  - `$h_{vgty}^*$` is the optimal harvest per trip.
  - `$T_{vgy}^*$` optimal total number of trips.
- Optimal quota level, per year and by species, is obtained by solving the social-planner optimization problem to maximize the net value
of the fishery